Download MECHANISMS OF CENTRAL TRANSMISSION OF RESPIRATORY

ACTA NEUROBIOL. EXP. 1973, 33: 287-299
Lecture delivered at Symposium "Neural control of breathing"
held in Warszawa, August 1971
MECHANISMS OF CENTRAL TRANSMISSION
OF RESPIRATORY REFLEXES
H. P. KOEPCHEN, D. KLUSSENDORF and U. PHILIPP
Institute of Physiology, Freie Universittit Berlin, West Berlin
Abstract. Several types of respiratory reflex actions can be discerned according
to the reactions of typical respiratory neurons in the efferent part of the central
rhythmogenic structure. Whereas respiration runs closely parallel with inspiratory
neuron activity the behaviour of expiratory neurons cannot be derived from the
resulting reflex changes of respiration. So expiratory apnoea can be combined
with continuous activity or with inactivation of expiratory neurons. Extracellular
records from a closed uniformly reacting population of expiratory neurons and
from neighbouring reticular neurons allowed experimental differentiation between
different types of central respiratory reflex actions. In experiments on anaesthetized
dogs the responses to chemoreceptor and baroreceptor excitation and to pulmonary
inflation were investigated. Chemoreceptor excitation leads to activation of inspiratory, expiratory, and reticular neurons, whereas the baroreceptor afferents act
in the opposite direction. In contrast moderate lung inflation causes more specific
effects: activation of expiratory neurons, inactivation of inspiratory neurons. But
if a certain degree of lung inflations is exceeded a more general inhibition of both
inspiratory and expiratory neurons takes place. These results only apply to the
"typical" respiratory neurons. The principles used to distinguish between the different types of reflexes are proposed for a basis of classification also of other neural,
chemical or pharmacological influences on breathing.
The mode of central respiratory reflex transmission is still fairly
unknown. The peculiarity in comparison with other reflex centres is that
at least partly the same complicated network of brain stem neurons is
involved which also generates the respiratory rhythm. A s long as the
fundamental process of central rhythmogenesis is not fully understood,
also respiratory reflex transmission will be a partly unsolved problem.
In spite of this some general principles can be derived from the reactions
of certain groups of respiratory neurons.
288
H. P. KOEPCHEN ET AL.
Figure 1A shows a simple functional scheme of this structure in order
to demonstrate the different possibilities of respiratory reflex transinision. This scheme is based on the following fairly well established
facts. There are two populations of respiratory neurons, one inspiratory
the other expiratory, which are alternately active and have mutual reciprocal inhibitory connexions. The inspiratory neurons partly activate
-
-
INSPIRATORY
-
EXPIRATORY
POMATION
POWLATION
+
-
,
t
A+
UNSPECIFIC
RETICKAR NEURONS
INFLATION
Fig. 1. Schematic drawing of the basic functional structure i n typical brain stem
respiratory and unspecific reticular neurons (A) and the probable mode of action
of various types of respiratory reflex afferents on this system (B).The arrows
from above represent the influences from higher structures of the neuraxis, and
the circles represent the facilitatory and inhibitory feedback mechanisms within
the respiratory populations, Interrupted line: effect of strong and/or rapid lung
inflation. It should be noted that besides the drawn indirect actions via unspecific
general activation or inactivation, direct connexions from chemoreceptor and baroreceptor afferents to inspiratory and expiratory neurons are not excluded by the
experiments. The given scheme is the simplest possible interpretation of the
present findings.
inspiratory motoneurons thus causing inspiration, the expiratory neurons
discharge during the non-inspiratory phase and at least some of them
can activate expiratory motoneurons. The activity in this structure is
closely related to general reticular activity (Salmoiraghi and Burns 1960)
and activation of respiratory neurons can be one part of a general reticular activation (Hugelin and Cohen 1963).
The following three types of respiratory reflexes will be considered:
(i) those increasing respiration, (ii) those depressing respiration, and (iii)
those shifting the relation between inspiratory and expiratory activity.
Because these reflexes act on the central rhythmogenic apparatus sketched before several types of reflex mechanisms are1 possible (Table I).
CENTRAL TRANSMISSION O F REFLEXES
389
Compilation of the different possible reactions of typical inspiratory, expiratory, and reticular
neurons underlying driving and depressing respiratory reflexes
Reflexes driving respiration
Typical
inspuatory
neuron
Typical
expiratory
neuron
"Unspecific"
reticular
neuron
I
Reflexes depressing respiration
Typical
inspiratory
neuron
1
Typical
expiratory
neuron
I
1
"Unspecific"
reticular
neuron
Reflexes driving respiration can operate by:
a) Primary activation of inspiratory neurons combined with reciprocal
secondary inhibition of expiratory neurons.
b) Primary inhibition d expiratory neurons which in time will lead
to disinhibitioln of inspiratory neurons.
c) General activation of reticular activity causing increased firing of
inspiratory as well as expiratory neurons.
d) Specific activation of inspiratory as well as expiratory neurons
without involvement of "unspecific" reticular neurons.
By means of extracellular recordings from respiratory and reticular
neurons only the type a and b respiratory drive can be distinguished
from c and from d, but not a from b. For reflexes depressing respiration
the same considerations are applicable but with the inverse sign (right
part of Table I). Table I also makes clear that records from inspiratory
neurons cannot help to differentiate between the different types of reflex
activation or depression of respiration. Therefore we have concentrated
our studies mainly on expiratory neurons.
Another consequence of the functional structure shown in the upper
part of Fig. 1 is that two kinds of respiratory arrest in expiratory position
are possible which cannot be differentiated by recording inspiratory
impulses or respiratory movements: (i) The general activity in the system can be diminished to such a degree that respiratory neurons cease
to fire all together. We have called this state "inactive apnoea". (ii)
Expiratory neurons can be continuously active while only inspiratory
activity ceased, which is called ,,arrhythmic apnoea" (Koepchen et a1
1970) (Fig. 2). Arrhythmic apnoea gives an opportunity to study direct
actions on expiratory neurons apart from responses caused indirectly by
way of action on the inspiratory population (qee below).
19
- Acta
Neurobiologiae Experimentalis
Succ~nylchol~ne
Art~tlclalventllat~on
Chlorolosr
i. v.
AOlliC
pressure
1'00
~~~~~~~~~~~~~~~~~mw~~~~~~~~~~~~~~~\~\~~~~~\
Integrated d i s h a m
01 phrenic nerve
Fig. 2. Time course of multiple neuronal activity in the centre of the typical expiratory population during generation of arrhythmic apnoea induced by an overdose
of chloralose. Pith: intrathoracic pressure recorded by means of a balloon in the
oesophagus. Inspiration goes upwards during artificial ventilation after immobilization with succinylcholine.
The scheme of Table I may be valid only for the efferent part of
the system whereas in the proper central interneurons other kinds of
reaction are conceivable. f i r the efferent part of the inspiratmy neurons
phrenic nerve activity is a good indicator. A closed population of expiratory neurons was localized by Philipp and Klussendorf (1969) in the
dog caudo-lateral to the obex. This localization roughly agrees with
statements of other authors for the cat (Baumgarten et al. 1957, Haber
et al. 1957, Nelson 1959, Batsel 1964, Merrill 1970, Bianchi 1971) but in
the dog this expiratory area is still more extensive and sharply limited
against the surrounding non-respiratory reticular regions. Within this
area all neurons react in the same manner in the course of respiratory
reflexes as in the example of Fig. 3. Here during the inhibition caused
by blood pressure increase the activity d one single expiratory neuron
js separated by electronic means from the background activity recorded
with the same electrode. As can be seen the single neuron activity runs
closely parallel with that in the surrounding expiratory population.
Stimulation through the recording electrode in the centre of this
region leads to expiratory arrest of respiration (Fig. 4). Stronger stimuli
cause forced expiratory movements. On the other hand the burst activity
CENTRAL TRANSMISSION OF REFLEXES
Carotid sinus distension
Aortic occlusion
Aortic occlusion
m
291
I
I
Aor bc
pressure
rnull~plerecord
..stondord"
octaon pacenttols
of rnclllple record
Fig. 3. Neuronal activity in the centre of the typical expiratory population from
several neighbouring neurons (lower record) recorded with the same microelectrode
during reflex changes caused by baroreceptor excitation. At the same time activity
of one single neuron is separated from the background of multiple activity by
electronic means (upper record). The reflex changes of discharge frequency run
closely parallel in the multiple and single unit records demonstrating the uniformity of reaction in the whole recorded population.
within this area is diminished after stimulation in the inspiratory region
adjacent rostrally. Therefore it seems to be justified to call these neurons:
"typical expiratory neurons". Their function is (i) to cause expiration,
and (ii) to inhibit inspiratory activity. The following analysis of reflex
effects is based on this special group of expiratory neurons in dogs
anaesthetized with chloralose-urethane.
As a prototype for activating reflexes the chemoreceptsr reflexes
were investigated, because there are contradictory findings concerning
the chemoreflex action on expiratory neurons (Baumgarten 1956, Batsel
1965, Nesland et al. 1966). We found that chemoreceptor stimulation
caused by i.v. injection of lobeline (0.05 mgkg) activated both inspiratory
and expiratory neurons (Fig. 5). The impulse pattern was analysed by
computer: during chemoreceptor stimulation the mean frequency of impulses during the bursts and the number of impulses/time were increased
Sl~mulal~on
through
the rrcordnng
microrlrclrodc
- r
Aortic
pressure
Respiration
(Spirometer)
Integrated
multiple
rxplmtory activity
3 min
1
Insp.
0
10
20
30
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
SCC
Fig. 4. Record from the centre of the typical expiratory population (left), and effect of stimulation a t the same site through
the microelectrode used before as recording electrode (right). Respiration registrated by the aid of a spirometer and
simultaneously by a baloon in the oesophagus. Total arrest of respiration in expiratory position during the stimulation period.
CENTRAL TRANSMISSION O F REFLEXES
2 93
in both populations whereas the duration of bursts diminished in expiratory neurons.
It might have been possible that the activation of expiratory neurons
during chemoreceptor stimulation was the expression of some kind of
"rebound" after release from a stronger inhibition in the inspiratory
phase caused by the augmentation of inspiratory bursts. To test this
hypothesis we applied the chemoreceptor stimulus during the state of
Lobeline 1 mg i.v.
lntegroted
Exp~ratoryActivity
Standardized
Integrated
lnsplrotory Activity
Fig. 5. Simultaneous multiple records with two microelectmdes, one from the
expiratory region the other one from the ventral inspiratory region at the level
of the obex. Both populations are activated after i.v. injection of a low dose
of lobeline, which was ineffective after denervation of the peripheral chemoreceptom.
expiratory arrhythmic apnoea produced by a higher dose of chloralose
(Fig. 6). In this instance also expiratory neurons were activited by
chemore~epto~r
excitation in the absence of inspiratory activity. That
means that chemoreceptor afferents excite expiratory neurons independently of their action on inspiratory neurons. In other experiments the
response of surrounding non-respiratory reticular neurons to chemore-
H. P. KOEPCHEN ET AL.
Apnea by chloralose
Lobeline
Aortic occlusion
lntcqroted
multiple
expiralory activily
1
0-
I
1I
,I
2
min
Aortic O C c l h
+-I
Lobcline
1.5 mg i.v,
Fig. 6. Reflex effects on the integrated typical multiple expiratory activity during
arrhythmic apnoea induced by chloralose. Artificial ventilation by a Starling pump.
Blood pressure increase during aortic occlusion depresses neuronal activity until
complete arrest whereas chemoreceptor stimulation by lobeline restores the expiratory activity in spite of the elevated blood pressure.
ceptor stimulation was recorded. For most of the tested reticular neurons
chemoreceptor excitation likewise led to increased activity (P. Langhorst
and H. P. Koepchen, unpublished data). Therefore the chemoreceptor
reflex increase of breathing is a generally activating reflex according to
case c in Table I .
The known inhibition of respiration by arterial baroreceptor afferents
(Heymans and Bouckaert 1930) was investigated as an example of an
inhibitory respiratory reflex. As was expected inspiratory neurons were
inhibited during the reflex hypopmea of baroreceptor origin. The effect
on the typical expiratory neurons is more complicated: the decrease of
respiratory frequency is accompanied by prolonged expiratory bursts.
Accordingly the number of impulses per burst is augmented, but on the
other hand the discharge frequency during these prolonged expiratory
bursts is diminished. Similar findings have been described in the cat
(Gabriel and Seller 1969), but the question remained unsettled if these
changes in the discharge pattern of expiratory neurons under the influ-
CENTRAL TRANSMISSION OF REFLEXES
295
ence of the baroreceptor input may be interpreted as activation or inhibition. More conclusive results can be obtained if the decrease of respiratory frequency during the blood pressure increase is prevented by artificial ventilation. In this case all parameters of expiratory burst activity
decrease (Kliissendorf et al. 1970). This inhibition also can be obtained if
the blood pressure during baroreceptor excitation is held constant (Fig. 7).
Ah in the case of chemoreceptor stimulation the baroreceptor reflex
Succinylcholtne
Arttf~clalvent~lat~on
Carotid sinus distens~on Aortic O C C I U S I O ~
Stondordized
octlon
lfi,s,~~~~~\h~w~n\\~h\~\t~\\k\~k~\\~\-\b\\\~:hh\~k~~\*~--
n;m ~g
Aortic
pressurr
~
~
~
~
,
,
l
I
v
~
~
~
~
bM\[\:&N,
lntegroted O C ~ ~ V I ~ Y
of slngle
explrolory neuron 0
I
60
30
I
I
I
I
1
1sec
Fig. 7. Effect of baroreceptor stimulation on single neuron activity in the typical
expiratory region during constant .respiratory frequency triggered by artificial
ventilation in the immobilized dog. Both carotid sinuses were distended by intraluminal balloons and aortic pressure was adjusted by gradual inflation of an
intra-aortic balloon. Note the immediate drop of discharge frequency after the
beginning of carotid sinus distension. The following reflex decrease of blood
presswe is accompanied by reactivation o f the expiratory neuron. At this moment
arterial pressure was brought to the control level by partial aortic conclusion.
The restoration of blood pressure depresses again the expiratory activity, which
remains lowered as long as carotid sinus distension is maintained.
effect on expiratory neurons is still present during arrhythmic expiratmy
apnoea (see Fig. 6), and therefore cannot be a secondary consequence of
primary action on the inspiratory population.
In earlier work we found that 75O/o of the reticular neurons in the
same region of the brain stem reacting to blood pressure changes are
296
H. P. KOEPCHEN ET AL.
inhibited (Koepchen et al. 1967). Thus the baroreceptor reflex can be
classified as a generally inhibiting reflex and is the mirror-image of
chemoreflex excitation (see column c in Table I).
The most important examples of reflexes changing the balance between inspiPutory and expiratory activity are the mechano-reflexes from
the lungs (Hering-Breuer reflexes). Some details of the central reflex
action of pulmonary inflation and deflation have been described elsewhere (Kliissendorf and Koepchen 1969, Koepchen et al. 1971). In a limited range of lung volume changes we could confirm in the dog the excitation of expiratory neurons and the inhibition of inspiratory neurons
described in experiments on cats and rabbits (Hukuhara et al. 1956,
Baumgarten and Kanzow 1958, Koepchen and Baumgarten 1958, Nesland
and Plum 1965, Cohen 1969).
But we found some remarkable pecularities in the response of expiratory neurons to lung distension which will be discussed here in so far
as they are related to the classification of pulmonary distension reflexes
in the scheme mentioned above.
Records from the centre of the typical expiratory neuron population
Start Lung Inflation
1
Pressure
loo
h ! ~ ~ d ~ , f i M ~ \ + ~ + ~ w w w ~ ~ ~ t d ~ ~ ~ h ~ \ & - .
lnlegroted
Expiratory A c l ~ t y
Fig. 8. Effect of lung inflation on multiple activity in the expiratory population.
Stepwise increase of pulmonary distension produced by continuous blowing of OP
into the open tracheal tube causes a stepwise increase of the level of expiratory
activity together with the prolongation of the bursts. Further increase of intrapulmonary pressure (at the right) depresses the activity of the same neurons.
If the overdistension is somewhat reduced, increased expiratory discharge frequency reappears.
297
CENTRAL TRANSMISSION OF REFLEXES
showed that during gradual inflation of the lungs the same neurons
whose discharge frequency increased during moderate inflation were
inhibited at higher degrees of inflation (Fig. 8). This inhibition of expiratory neurons was observed even a t moderate lung volumes if the rate
of inflation was increased. The secondary inhibition of expiratory neurons at higher lung volumes is not accompanied by inspiratory efforts or
phrenic nerve activity and therefore cannot be related to the Head's
(1889) "Paradoxical reflex". That means that under certain conditions
expiratory, as well as inspiratory activity is depressed by pulmonary
inflation. At these higher ranges of lung inflation we could observe also
inhibition of unspecific reticular activity (Koepchen 1969, Fig. 8).
These findings suggest that two components of the lung inflation
reflex may be discerned: the firs,t one shifting the balance between
central inspiratory and expiratory activity in favour of the latter, and
the other one causing general central depression analogous to the baroreceptor reflex (see Table 11). We have no information yet as to whether
these two components are due to the action of different groups of pulmonary receptors or to the different response of expiratory neurons dependent on the frequency in the afferent vagal fibres.
In arrhythmic states (chloralose apnoea and hyperventilation apnoea)
we never could observe the activation of the typical expiratory neuron
population during moderate lung inflation, which was present in the
same neurons in the case of intact central respiratory rhythm. Only
inhibition was achieved by all degrees and rates of lung inflation during
the arrhythmic states.
Schematic representation of the results obtained in the reflex studies on respiratory and reticular
neurons. The responses to baroreceptor stimulation, chemoreceptor stimulation, and strong and/or
rapid lung inflation correspond to the type c reaction in Table I, i.e. general activation or inactivation of reticular activity
Inspiratory neuron
Rhythmic
state
Chemoreceptor
stimulation
II t / I
Baroreceptor
stimulation
Moderate
and slow
,
P~llmonaryinflation 1
Strong and/or
rapid
4
Expiratory neuron
Rhythmic
state
Arrhythmic
state
i
.
1
'
Arrhythmic
state
1
.
1
I
Reticular
neuron
.
1
2 98
H. P. KOEPCHEN ET AL.
The scheme in Table I1 and Fig. 1 comprise the findings and give the
basis for the suggested classification of the studied respiratory reflexes.
It should be noted that our present data on expiratory n e u m s are much
more complete than those on inspiratory neurons. Of course such a sim-'
ple classification in the first instance pertains only to "typical" respiratory neurons representing the efferent part of the rhythmogenic system.
The closed expiratory population described above probably belongs to
this efferent part, just as those inspiratory neurons reacting in the same
manner are related to phrenic nerve activity.
Bearing in mind this limitation it would be useful to consider also
other nervous, chemical and pharmacological influences on the respiratory centres with regard to their position in the outlined scheme.
This investigation was supported by the Deutsche Forschungsgemeinschaft.
REFERENCES
BATSEL, H. L. 6964. Localization of bulbar respiratory center by microelectrode
sounding. Exp. Neurol. 9: 410-426.
BATSEL, H. L. 1965. Some functional properties of bulbar respiratory units. Exp.
Neurol. 11 : 341-361.
BAUMGARTEN, R. von 1956. Koordinationsformen einzelner Ganglienzellen der
rhombencephalen Atemzentren. Pfliig. Arch. Ges. Physiol. 262: 573-594.
BAUMGARTEN, R. von, BAUMGARTEN, A. von and SCHAEFER, K. P. 1957.
Beitrag zur Lokalisationsfrage bulboreticularer respiratorischer Neurone der
Katze. Pflug. Arch. Ges. Physiol. 264: 217-227.
BAUMGARTEN, R. von and KANZOW, E. 1958. The interaction of two types of
inspiratory neurons in the region of the tractus solitarius of the cat. Arch.
Ital. Biol. 96: 361-373.
BIANCHI, A. L. 1971. Localisation et etude des neurones respiratoires bulbaires.
Mise en jeu antidmmique par stimulation spinale ou vagale. J. Physiol.
(Paris) 63: 5-40.
COHEN, M. I. 1969. Discharge pattern of brain-stem respiratory neurons during
Hering-Breuer reflex evoked by lung inflation. J. Neurophysiol. 32: 356-374.
GABRIEL, M, and SELLER, H. 1969. Excitation of expiratory neurones adjacent
to the nucleus ambiguus by carotid sinus baroreceptor and trigeminal aff e r e n t ~ Pfliig.
.
Arch. Ges. Physiol. 313: 1-10.
HABER, E., KOHEN, K. W., NGAI, S. H., HOLADAY, D. A. and WANG, S. C.
1957. Localization of spontaneous respiratory neuronal activities i n the medulla
oblongata of the cat: A new location of the expiratory center. Amer. J.
Physiol. 190: 350-355.
HEAD, H. 1889. On the regulation of respiration. J. Physiol. (Lond.) 10 (1): 279-290.
HEYMANS, C. and BOUCKART, J. J. 1930. Sinus caroticus and respiratory reflexes.
J. Physiol. (Lond.) 69: 254-266.
HUGELIN, A. and COHEN, M. I. 1963. The reticular activating system and respiratory regulation in the cat. Ann. N. Y. Acad. Sci. 109: 586-603.
CENTRAL TRANSMISSION OF REFLFXES
299
HUKUHARA, T., OKADA, H. and NAKAYAMA, S. 1956. On the vagus respiratory
reflex. Jap. J. Physiol. 6 : 87-97.
KLUSSENDORF, D. and KOEPCHEN, H. P. 1969. Automatic analysis of discharge
patterns of respiratory neurons of dog during respiratory reflexes from
baroreceptor, chemoreceptor and pulmonary afferents. Pfliig. Arch. Ges.
Physiol. 312: 58.
KLUSSENDORF, D., PHILIPP, U. and KOBPCHEN, H. P. 1970. Studies on the
central mechanism of reflex inhibition of respiration by baroreceptor aff e r e n t ~ .Pfliig. Arch. Ges. Physiol. 319: R50.
KOEPCHEN, H. P. 1969. Vegetative-somatic relationships in single neurone activity
in the lower brain stem. In R. C. Evans and T. B. Mulholland (ed.), Attention
in neurophysiology. Buttenvorths, London, p. 83-99.
KOEPCHEN, H. P. and BAUMGARTEN, R., von 1958. Entladungsmuster und vagale
Beeinflussung exspiratorischer Neurone im Hirnstamm der Katze. Pfliig.
Arch. Ges. Physiol. 268: 64. (Abstr.).
KOEPCHEN, H. P., KLUSSENDORF, D., BILAN, M., SAPUNAROW, N. and SOMMER, D. 1971. Central transmission of pulmonary inflation and deflation
reflexes. Proc. XXV Int. Congr. Physiol. Sci. (Munich) 9: 312.
KOEPCHEN, H. P., KLUSSENDORF, D. and PHILIPP, U. 1970. The discharge
pattern of expiratory neurons during various states of apnea. Pfliig. Arch.
Ges. Physiol. 319: R51.
KOEPCHEN, H. P., LANGHORST, P., SELLER, H., POLSTER, J. and WAGNER,
P. H. 1967. Neuronale Aktivitat im unteren Hirnstamm des Hundes mit
Beziehung zum Kreislauf. Pfliig. Arch. Ges. Physiol. 294: 40-64.
MERRILL, E. G. 1970. The lateral respiratory neurones of the medulla: their associations with nucleus ambiguus, nucleus retroambigualis, the spinal accessory
nucleus, and the spinal cord. Brain Res. 24: 11-28.
NELSON, J. R. 1959. Single unit activity i n medullary respiratory centers of cat.
J. Neurophysiol. 22 : 590-598.
NESLAND, R. and PLUM, F. 1965. Subtypes of medullary respiratory neurons. Exp.
Neurol. 12: 337348.
NESLAND, R. S., PLUM, F., NELSON, J. R., and SIEDLER, H. P. 1966. The graded
response to stimulation of medullary respiratory neurons. Exp. Neurol. 14:
57-76.
PHILIPP, U. and KLUSSENDORF, D. 1969. Localization and demarcation of a population of neurons active a s a whole during expiration in the lower brain
stem of dog. Pfliig. Arch. Ges. Physiol. 312: 57.
SALMOIRAGHI, G. C. and BURNS, B. D. 1960. Notes on mechanism of rhythmic
respiration. J. Neurophysiol. 23: 14-26.
H. P. KOEPCHEN, D. KLUSSENDORF and U. PHILIPP, Physiologisches Institut der Freie
Universitat Berlin, 1 Berlin 33 (Dahlem), Arnimallee 22, West Berlin.